From owner-freebsd-hackers Wed Sep 18 23:29:57 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id XAA21897 for hackers-outgoing; Wed, 18 Sep 1996 23:29:57 -0700 (PDT) Received: from casparc.ppp.net (casparc.ppp.net [194.64.12.35]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id XAA21800 for ; Wed, 18 Sep 1996 23:29:44 -0700 (PDT) Received: from ernie by casparc.ppp.net with uucp (Smail3.1.28.1 #1) id m0v3cBh-000HzcC; Thu, 19 Sep 96 08:02 MET DST Received: by ernie.kts.org (Smail3.1.29.1 #5) id m0v3bgB-00000bC; Thu, 19 Sep 96 07:29 MET DST Message-Id: From: hm@kts.org (Hellmuth Michaelis) Subject: Re: Special Cycles on the PCI bus To: dg@root.com Date: Thu, 19 Sep 1996 07:29:39 +0200 (MET DST) Cc: durian@plutotech.com, msmith@atrad.adelaide.edu.au, freebsd-hackers@FreeBSD.org In-Reply-To: <199609181626.JAA12461@root.com> from "David Greenman" at Sep 18, 96 09:26:50 am Organization: Kitchen Table Systems Reply-To: hm@kts.org X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk David Greenman wrote: > Perhaps they are caused by the inb(0x84)'s that are used as time delays in > a couple places of the kernel. The inb(0x84) is supposed to be an unused port > and thus should take about 1.25us to read from. This isn't always true, > however, as people have pointed out in the past. We should probably replace > the last few places where these are used with a calibrated wait loop. And then the old question appears again - how to do (relatively) precise delays in the range of 1 us ... 10 us. As far as i understood Bruce that time, DELAY isn't able to handle this precisely. Has that changed ? hellmuth -- Hellmuth Michaelis hm@kts.org Hamburg, Europe (A)bort, (R)etry, (I)nstall BSD ?